Post on 22-May-2020
Working Paper Series No 53 / August 2017
Two big distortions: bank incentives for debt financing
by Jesse Groenewegen Peter Wierts
ABSTRACT
Systemically important banks are subject to at least two departures from the neutrality of debt versus
equity financing: the tax deductibility of interest payments and implicit funding subsidies. This paper fills
a gap in the literature by comparing their mechanism and interaction within a common analytical
framework. Findings indicate that both the tax shield and implicit funding subsidy remain large, in the
order of up to 1 percent of GDP, despite decreases in recent years. But the underlying mechanisms
differ. The tax shield incentivises debt financing as it reduces tax payments to the government. The
implicit funding subsidy incentivises debt financing as it lowers private bankruptcy costs. This funding
subsidy is passed on to other bank stakeholders. It therefore provides incentives for increases in balance
sheet size and risk taking. This, in turn, increases the value of the tax shield. Overall, these results help
to explain why systemically important banks are highly leveraged.
JEL codes: G21, G32, H25
Keywords: taxation, subsidies, debt, leverage.
1
1 Introduction
Every standard textbook in corporate finance covers Modigliani-Miller proposition 1 (MM-1): capital
structure does not influence firm value. And it will explain how tax deductibility of interest payments
leads to a departure from MM-1: more debt financing generally increases firm value. In the past, many
estimates have been made of the size of this so-called ‘tax shield’. More recent empirical research
shows that higher corporate income tax rates lead to more debt financing, both for banks and for the
corporate sector as a whole (e.g. de Mooij and Keen, 2016). Insofar as banks are affected, this increases
the likelihood of financial crises (de Mooij et al., 2014).
Banks differ from other firms in that they are subject to an additional departure from the neutrality of
debt versus equity financing: implicit funding subsidies for banks that are considered too-big-to-fail. The
literature includes empirical estimations of funding advantages, e.g. of up to 100 basis points (IMF,
2014) and discussions of possible effects on balance sheet structures (Morrison, 2011).
Until now, the literature has focused almost exclusively on these incentives for debt financing in
isolation. But the overall incentive effect for banks will be determined by the interaction of different
taxes and subsidies together; they may reinforce or offset each other depending on their size, direction
and base. This study aims to fill this gap in the literature. We analyse the sizes, mechanism and
interactions of the tax shield and the implicit subsidy in a common analytical framework.
Our main findings are the following. First, we find that the sizes of the tax shield and implicit funding
subsidy are large and have the potential to influence bank behaviour. The size of the tax shield lies in the
order of 1 to 2 percent of GDP in 2009-2010 and between 0.3 and 0.9 percent of GDP more recently, as
interest rates have declined. The implicit subsidy for too big to fail banks has also been substantial,
peaking around 2009 and 2012, and still amounts to nearly 1 percent of GDP in some countries in recent
years.
Second, the tax shield and implicit subsidy can both be integrated in standard trade-off theory. Theory
predicts that these incentives increase debt financing, respectively by increasing the value of the tax
shield and by lowering private bankruptcy costs. Debt financing increases firm value as it reduces the
amount of tax a bank pays to the government. The implicit subsidy provides an incentive for debt
financing given that bondholders demand a lower risk premium, as part of the default risk is shifted to
the government. This subsidy can be passed on to other debt holders (i.e. expanding debt capacity), to
asset holders (i.e. expanding the asset side; taking on riskier assets) or equity holders (i.e. directly
2
increasing return on equity). The two incentives also interact: as the implicit subsidy stimulates debt
financing, it further increases the value of the tax shield. As a result, both effects help to explain why
systemically important banks have an incentive to grow in size and remain highly leveraged.
Our paper builds on the (separate) strands in the literature on the role of taxation in banking and the
implicit subsidy on debt financing. To start with the former, the consensus is that corporate taxation,
and specifically the tax deductibility of interest payments, influences the financing decisions of
corporations.1 In recent years, research has looked at the effects of taxation on the capital structure of
financial firms (Gropp and Heider, 2009). Such firms were previously excluded from studies on corporate
taxation and capital structure because their financing decisions were assumed to depend mainly on
mandatory capital buffers and not on tax incentives. However, several recent empirical studies suggest
that taxation does affect bank leverage. Keen and de Mooij (2012), Heckemeyer and de Mooij (2013),
Langedijk et al. (2014), Hemmelgarn and Teichmann (2014) and Luca and Tieman (2016) all find
substantial effects of taxation on the financing structure of banks.
The literature finds that the implicit subsidy is also an important driver of banks’ decisions about their
capital structure. IMF (2014) and Morrison (2011) analyze the incentive effects for banks that are too big
to fail. They stress that the price of debt becomes relatively risk insensitive, due to the implicit
guarantee to debt holders. Banks therefore have an incentive to take on higher levels of cheap debt, and
invest it in riskier assets that benefit shareholders. In sum, IMF (2014) argues that too-big-to fail funding
subsidies reflect cheaper debt, and therefore lead to more leverage, more risk-taking, a larger balance
sheet, and more systemic risk to society as a result. Implicit subsidies on debt are also cited as a major
reason why banks are much more leveraged than regular, non-financial companies, although both have
the same theoretical financial benefits from the tax shield. Admati et al. (2015) argue that implicit
subsidies on debt soften creditors’ incentive to restrict leverage through covenants or high interest
rates. Moreover, once debt is in place, bringing back debt makes the remaining debt safer, and
therefore represents a transfer of wealth from shareholders to the government or debt holders. Existing
shareholders will therefore resist decreasing leverage once it is in place. Aside from taxes and explicit
and implicit guarantees, agency distortions can also play a role in incentivizing banks toward using more
debt financing (Admati and Hellwig, 2013).
1 For an overview of the literature, see Auerbach (2002) or Graham (2006).
3
The remainder of this paper is organized as follows. Section 2 looks at the size of the tax shield and
implicit funding subsidy, using a common denominator by expressing them as a percentage of GDP.
Section 3 compares the underlying mechanisms by which the tax shield and the implicit subsidy affect
finance structures. Section 4 puts our findings in a broader perspective.
2 Size
The tax shield can be measured as the increase in yearly cash flow due to tax advantages from debt
financing, relative to an unlevered firm.2 It should not be interpreted as a direct transfer of resources. If
the bank is profitable, it still pays taxes to the government, but the tax shield indicates that it reduces its
tax bill by using debt financing.3 Moreover, the tax shield does not measure the overall tax pressure in a
more complete framework. This would have to take into account the tax treatment of interest payments
on the receiving side as well. There are some studies which show that higher personal income tax rates
on interest relative to dividends and capital gains reduce leverage ratios (Lin and Flannery, 2013). But a
significant share of investments is sheltered from personal income taxes, such as those by institutional
investors who are tax exempt (IMF, 2016). Moreover, investors can find ways to escape taxes through
international tax arbitrage. In practice, the largest share of bank wholesale funding is held by
institutional investors.
The tax shield can be calculated by multiplying interest payments on debt with the applicable corporate
income tax rate. As an illustration, we do this for the banking sector as a whole of five countries: France,
Germany, Italy, the Netherlands and the United Kingdom (UK). Our sample spans the period 2009-2016.
The results are in Figure 1. See Box 1 for further information on data sources and our calculation
method.
2 By calculating the net present value of all future cash flows, it can also be expressed in terms of firm value.
3 On the other hand, if a bank is not profitable, it does not pay corporate income tax and hence it also does not benefit from the tax shield.
4
Figure 1. Size of the tax shield per banking sector (as a percentage of GDP)
Results indicate that the tax shield ranged from roughly 0.3 percent of GDP in Germany in 2016 to over
two percent of GDP in the United Kingdom in 2009. These results are broadly in line with estimations by
The Economist (2015), which arrives at an estimate of 0.7 percent of GDP for the financial sector in the
euro area. As expected, our results show that the size of the tax shield has declined over the past few
years due to lower interest rates and decreases in the size of banks.4 But its remains large, providing a
substantial stimulus for debt financing.
The implicit funding subsidy is the decrease in funding costs due to an implicit guarantee from the
government. The funding cost reduction follows from the expectation that the government would
provide support to the bank in case of a possible bankruptcy, due to the externalities of such an event to
the rest of the financial system. We use the same sample of countries for estimating the size of the
implicit subsidy, considering the period 2009-2016. Figure 2 shows substantial effects of the implicit
subsidy in terms of basis points. See Box 2 for further information on the data and the estimation
method.
4 To assign the tax shield to the proper country, we use interest payment data from the BIS locational banking statistics (See also Box 1).
Since these are interest payments data, it is difficult to say whether the decline in the tax shield is due to lower rates or smaller balance
sheets.
0.0%
0.5%
1.0%
1.5%
2.0%
2.5%
France Germany Italy Netherlands UnitedKingdom
2009 2010 2011 2012 2013 2014 2015 2016
5
Figure 2. Size of implicit funding subsidy per banking sector (in basis points)
A first peak appears in 2009. At that time, market sentiment was rather risk-averse, leading to exploding
implied yields for bonds with low investment ratings. A few banks in our sample were perceived by
markets as on the brink of insolvency in 2009. These banks were strongly supported (or de facto
nationalized) by their sovereigns. Thus, especially for this year, a significant gap emerged between the
stand-alone rating and the support rating. Moreover, we observe a second peak in 2012, at the height of
the sovereign debt crisis. This is followed be a strong decline in recent years, as systemic risk has
receded and possibly also due to the introduction of bail-in regimes.
Values for recent years generally match the values found in the literature on the implicit subsidy. IMF
(2014) estimates the funding advantage to be approximately 60-90 basis points in 2014. Germany and
France generally have slightly higher values than the Netherlands and the United Kingdom. The funding
advantage for Italian banks is relatively small due to the relatively low credit rating of the sovereign,
which mechanically translates into a low funding advantage. The implicit funding advantage of Italian
banks is therefore also low.
Finally, we calculate the size of the implicit subsidy as a percentage of GDP. See Figure 3. Calculation
details are again in Box 2.
0
100
200
300
400
500
600
700
France Germany Italy Netherlands UnitedKingdom
2009 2010 2011 2012 2013 2014 2015 2016
6
Figure 3. Size of implicit subsidy per banking sector (as a percentage of GDP)
Results as presented in Figure 3 show broadly the same time pattern as in Figure 2, and the same
relative differences across countries. By 2016, the estimated size of the implicit subsidy remains large, as
it ranges between 0.4 and 1% of GDP in most countries. Its size is of the same order of magnitude as the
tax shield estimates in Figure 1, confirming its relevance for affecting bank behaviour.
0.0%
1.0%
2.0%
3.0%
4.0%
5.0%
6.0%
7.0%
8.0%
France Germany Italy Netherlands United Kingdom
2009 2010 2011 2012 2013 2014 2015 2016
7
Box 1. Data sources and calculation method – Tax shield
For calculating the tax shield, we use consolidated banking data from the ECB, supplemented with
data on interest payments from BIS locational banking statistics. Data on CIT rates are from the
OECD tax database. Interest payments are usually reported at the consolidated level but are
deductible against taxes paid nationally, which poses a challenge. To illustrate this issue, consider a
bank with headquarters in country A that has a subsidiary in country B. While the bank’s interest
payments are reported as consolidated in country A, taxes on the foreign subsidiary are levied by
the foreign authorities in country B at the level of this entity. Interest payments by this subsidiary,
therefore, are deductible against taxes paid in country B and should be attributed to the tax
advantage that country B provides to its banking sector.
As a result, the size of the tax subsidy should be calculated by taking the geographic orientation of
the banking system in each country into account. We solve this issue by using interest payments
data from BIS locational banking statistics that are at the entity level.
8
Box 2. Data sources and calculation method – implicit subsidy
The implicit funding subsidy cannot be observed directly, but needs to be estimated. Different
methods have been used in the literature for doing so, which all have their pros and cons (Kroszner,
2013; see Schich and Lindh, 2012, and Ueda and Weder di Mauro, 2013, for different approaches).
In our approach, we compare the ‘standalone ratings’ for a sample of large banks for which ratings
are available with the ‘support ratings’ that include the expected government support. For our
calculations of the implicit funding subsidy, we use Moody’s Long Term Issuer Credit Ratings and
Baseline Credit Assessment (BCA) Ratings from SNL Financial.
Subtracting the spread associated with the Long Term Issuer Credit Rating (support rating) from the
spread associated with Moody’s Baseline Credit Assessment Rating (standalone rating) in a given
year yields the funding cost advantage for a given bank in that year. We average the yearly funding
cost advantages for large banks in each country to arrive at annual estimates of the implicit funding
subsidy per country. We do this for a representative sample of 18 major banks in the
aforementioned five European countries for which ratings are available. We derive the average
annual funding costs for those ratings by using Merrill Lynch bond spread indices from Bloomberg.
As rating agencies started providing separate stand-alone and support ratings a few years ago, our
sample spans the years since 2009.
We also have to estimate the ratings-sensitive funding base of large banks. As also mentioned by
Noss and Sowerbutts (2012), some judgement is involved in determining which part of the debt base
can be considered as ratings-sensitive. As stated above, it is commonly understood that debt at
shorter maturities will be less sensitive than longer term debt. For our estimation, we include 50
percent of deposit funding (as about 50 percent of deposits are guaranteed under the deposit
guarantee system) and 85 percent of regular securities issued (as about 15 percent of securities has
a maturity shorter than one year). Moreover, we adjust for the share of the five largest banks in
each country, as a proxy for systemically important banks. By multiplying this funding base with the
average annual funding cost advantage per country we arrive at an estimate in euros of the implicit
subsidy, which can be compared with the nominal size of GDP for the countries in our sample.
9
3 Mechanism
According to the trade-off theory of capital structure, firms weigh the present value of interest tax
shields against the costs of financial distress, which both increase with debt financing (Myers, 1984):
Value of firm=value if all equity financed + PV(tax shield) - PV(costs of financial distress)
Implicit funding subsidies for too-big-to fail banks reflect risk shifting to the government, as reflected in
the funding subsidy, and hence lower private bankruptcy costs. This predicts that systemically important
banks will use a high degree of debt financing, and more debt financing than banks that are not
systemically important. The European Systemic Risk Board (2015, p. 19) provides anecdotic evidence for
both points. On average globally systemically important banks operated at 96.3 per cent debt financing
at the end of 2013, somewhat higher than 95.5 per cent of average debt financing for banks that are not
systemically important.5
In studying the underlying incentive mechanisms, we start from the standard assumption that firms
maximise their overall value, and hence their cash flows, which are assumed to be perpetual for
simplicity. We also assume perfect information and efficient markets, and no agency costs. The starting
point is a neutral benchmark, in which interest payments are non-deductible, and no implicit subsidy
exists.
Setup
We start with the most basic balance sheet, where A is total assets, E is equity,
𝐷𝑠𝑡 is short-term debt and 𝐷𝑙𝑡 is long-term debt, see Table 1. Short-term debt has a maturity of less
than one year and long-term debt has a maturity of more than one year. We assume that the funding
cost advantage does not apply to short-term debt. The probability of default in a period of one year or
less is very low, and thus independent of the systemic relevance of the bank (Bijlsma and Mocking,
2013).
5 ESRB (2015) recommends higher leverage constraints for systemically important banks, as the negative externalities from their distress or failure are more severe than for other banks.
10
Table 1. Basic bank balance sheet
Assets Liabilities A E 𝐷𝑠𝑡
𝐷𝑙𝑡
Assume a bank’s gross earnings on its assets A (i.e. before interest and taxes) are x. The interest rate on
short-term and long-term debt funding is 𝑟𝑠𝑡𝑑 and 𝑟𝑙𝑡
𝑑 respectively, where 𝑟𝑠𝑡𝑑 < 𝑟𝑙𝑡
𝑑, and the corporate tax
rate on earnings is t. Under a neutral tax regime, earnings, or x, would be taxed in their entirety first and
then redistributed to the holders of E, 𝐷𝑠𝑡 and 𝐷𝑙𝑡.
The cash flow to the bank’s financiers then consists of these parts:
To debt holders: 𝑟𝑠𝑡𝑑 𝐷𝑠𝑡 and 𝑟𝑙𝑡
𝑑𝐷𝑙𝑡
To shareholders, as residual claimants: (1 − 𝑡)𝑥 − 𝑟𝑠𝑡𝑑 𝐷𝑠𝑡 − 𝑟𝑙𝑡
𝑑𝐷𝑙𝑡
Total after-tax earnings can be calculated by adding up the different parts of the cash flow. They are:
(1 − 𝑡)𝑥 [1]
The bank cannot increase the total cash flow to its financiers by changing the debt equity mix in the
funding structure.
Deductibility of interest payments
We now assume that interest payments can be deducted from earnings before corporate income tax is
levied. This implies that the tax base will decrease when more debt is used instead of equity. The cash
flows to a bank’s financiers are:
To debt holders: 𝑟𝑠𝑡𝑑 𝐷𝑠𝑡 and 𝑟𝑙𝑡
𝑑𝐷𝑙𝑡
To shareholders E: (1 − 𝑡)(𝑥 − 𝑟𝑠𝑡𝑑 𝐷𝑠𝑡 − 𝑟𝑙𝑡
𝑑𝐷𝑙𝑡)
As before, adding up these parts gives the after-tax earnings:
(1 − 𝑡)(𝑥 − 𝑟𝑠𝑡𝑑 𝐷𝑠𝑡 − 𝑟𝑙𝑡
𝑑𝐷𝑙𝑡) + 𝑟𝑠𝑡𝑑 𝐷𝑠𝑡 + 𝑟𝑙𝑡
𝑑𝐷𝑙𝑡
= (1 − 𝑡)𝑥 + 𝑡𝑟𝑠𝑡𝑑 𝐷𝑠𝑡 + t𝑟𝑙𝑡
𝑑𝐷𝑙𝑡 [ 2 ]
11
Compare this expression to after-tax earnings in the absence of interest payment deductibility, see
equation [1]. The last two, additional, terms are the tax shield, or reduction in taxable income for an
individual bank. Given that debt D is a choice variable, the bank can reduce its tax payments, and
increase its after-tax earnings, by increasing D. In sum, taxes take away the irrelevance of capital
structure and a firm benefits from increasing leverage.
Implicit subsidy
The implicit subsidy works differently. Funding costs are lower, as bondholders require a lower risk
premium due to risk shifting to the government. If the bank would not react, equity holders would
benefit by getting a higher return. As a result, given gross earnings would be distributed differently: less
to bondholders and more to equity holders. Given that these flows cancel out, there would be no effect
on overall cash flows. However, this result would differ if returns on equity are taxed, as in our previous
case. The overall cash flow would now be lower, as a shift in return from bondholders to equity holders
increases the tax bill.
The bank therefore has an incentive to use the funding subsidy in a different manner. It can use part of
this funding advantage to increase the return to debt holders, increase debt funding and thereby
expand its balance sheet. Likewise, the bank can also pass on part of the subsidy to its customers,
cheapening the cost of mortgages, other bank loans or trading activities relative to its competitors, and
thereby attracting more business. Both effects would imply a debt-funded increase of the total balance
sheet.
We now include these effects in the same framework as before. As indicated, the implicit subsidy is
assumed to apply only to long-term debt. We denote the implicit subsidy (effectively a rate reduction on
long-term bank debt) by 𝑟𝑠. This implicit subsidy gives systemically important banks a funding advantage
over banks that are not systemically important, i.e. the required return on long-term debt is lower.
The banks can choose to pass on the funding advantage to different stakeholders. The implicit subsidy is
used to increase the return to short-term debt (with a share of 𝛼1), long-term debt (𝛼2), or decrease the
rates charged on assets (𝛼3). Equity holders get the residual share 𝛼4, with 𝛼1 + 𝛼2 + 𝛼3 + 𝛼4 = 1.
Subscripts s indicate that the subsidized bank will have more assets, more debt and hence higher gross
earnings, apart from the special case in which all the funding subsidy would be distributed immediately
to shareholders (i.e. 𝛼1 = 𝛼2 = 𝛼3 = 0).
12
The cash flow to the bank’s creditors is now:
To short-term debt holders: 𝑟𝑠𝑡𝑑 𝐷𝑠
𝑠𝑡 + 𝛼1𝑟𝑠𝐷𝑠𝑙𝑡
To long-term debt holders: 𝑟𝑙𝑡𝑑𝐷𝑠
𝑙𝑡 − 𝑟𝑠𝐷𝑠𝑙𝑡 + 𝛼2𝑟𝑠𝐷𝑠
𝑙𝑡
To asset holders 𝛼3𝑟𝑠𝐷𝑠𝑙𝑡
To shareholders: (1 − 𝑡)(𝑥𝑠 − 𝑟𝑠𝑡𝑑 𝐷𝑠
𝑠𝑡 − 𝑟𝑙𝑡𝑑𝐷𝑠
𝑙𝑡 + 𝛼4𝑟𝑠𝐷𝑠𝑙𝑡)
After tax earnings are:
(1 − 𝑡)𝑥𝑠 + 𝑡𝑟𝑠𝑡𝑑 𝐷𝑠
𝑠𝑡 + 𝑡𝑟𝑙𝑡𝑑𝐷𝑠
𝑙𝑡 − 𝑡𝛼4𝑟𝑠𝐷𝑠𝑙𝑡
This result, a larger balance sheet and therefore higher after-tax earnings, is in line with IMF (2014) and
Morrison (2011) who argue that that the overall effect of the implicit funding subsidy will be an increase
in the size of the balance sheet, and thereby an increase in gross earnings 𝑥𝑠. The final term is a
secondary effect: insofar the funding subsidy is passed on to equity holders it will be subject to taxation.
As indicated already, this provides an incentive to pass on the funding subsidy to debt and asset holders,
thereby enlarging the balance sheet. Moreover, as the bank holds more debt, the size of the tax shield
will also be larger. As a result, the implicit funding subsidy increases the size of the tax shield.
4 Conclusion
This paper has investigated the size of the tax shield and implicit subsidy, and their incentive effects. We
find that the sizes of the tax shield and implicit funding subsidy are large, and hence their potential to
influence bank behaviour is also large. The size of the tax shield lies in the order of 1 to 2 percent of GDP
in 2009-2010 and between 0.3 and 0.9 percent of GDP recently, as interest rates have declined. The
implicit subsidy for too big to fail banks has also been substantial, peaking around 2009 and 2012, and
still amounts to nearly 1 percent of GDP in some countries in recent years. Moreover, while both effects
provide an incentive for debt financing the underlying mechanisms differ.
The tax shield provides an incentive for debt financing since it reduces the amount of tax the bank pays
to the government. And the implicit subsidy provides an incentive for debt financing given that bond
holders demand a lower risk premium, as part of the default risk is shifted to the government. This
subsidy can be passed on to the other debt holders (i.e. expanding debt capacity), to asset holders (i.e.
expanding the asset side; taking on riskier assets) or equity holders (i.e. directly increasing return on
equity). As the implicit subsidy stimulates debt financing, it further increases the value of the tax shield.
13
As a result, both effects help to explain why systemically important banks have an incentive to grow in
size and remain highly leveraged.
There are several options for future research. Our paper has focused on the private costs and benefits of
debt financing only. One avenue for follow-up research would be to integrate these in a social
perspective. Excessive debt financing by banks produces negative external effects: e.g. increased
vulnerability to bank runs and amplification of asset price shocks. A social perspective would take these
broader, external costs into account.
Another option for future research is to integrate more debt incentives into a common framework. Our
research has started with the assertion that the overall incentive effects for debt financing follows from
the interaction of the individual taxes and subsidies. We have made a first attempt at an integrated
approach, but there are other incentives that we have not included. For example, we have focused on
the liabilities side only, ignoring asset-side taxation distortions such as those related to mortgage debt
and income tax on dividend payments. We also did not include the costs and benefits of bank levies,
deposit guarantee systems (DGS) and resolution funds.
A final avenue for research would be to explore policy options for neutralizing the debt bias, as capital
regulation and the incentives as discussed in this paper go in opposite directions. The options for a
more equal fiscal treatment of debt and equity financing are well-known (Bond, 2000; de Mooij and
Devereux, 2011). In addition, governments have instituted bank taxes in recent times to specifically
target bank leverage (Devereux et al., 2013). And different policies, including bail-in, have been
implemented to limit the implicit subsidy (FSB, 2011; DNB, 2015). An integrated view on the different
debt incentives requires an integrated policy approach, however. This would assure that key factors
affecting the financing choices of banks work in the same direction.
14
References
Admati, A.R. and M. Hellwig. 2013. The Banker’s New Clothes: What’s Wrong with Banking and What to do About it. Princeton University Press.
Admati, A.R., P.M. DeMarzo, M. Hellwig and P. Pfleiderer. 2015. The Leverage Ratchet Effect. Stanford Business School of Graduate Working paper 3029.
Auerbach, A.J. (2002). Taxation and Corporate Financial Policy. Handbook of Public Economics.
Bijlsma, M.J. and R.J.M. Mocking. (2013). The Private Value of Too-big-to-fail Guarantees. CPB Discussion Paper 240.
Bond, S.R. 2000. Levelling Up or Levelling Down? Some Reflections on the ACE and CBIT Proposals, and the Future of the Corporate Tax Base. in Taxing Capital Income in the European Union, S. Cnossen (ed.), Oxford: Oxford University Press.
Devereux, M.P., N. Johanessen and J. Vella. 2013. Can Taxes Tame the Banks? Evidence from the European bank levies. Oxford University Centre for Business Taxaxtion Working Paper 13/25.
DNB. 2015. Overview of Financial Stability, Spring 2015, Chapter 3: Ending Too-big-to Fail? De Nederlandsche Bank, Amsterdam.
ESRB. 2015. The ESRB Handbook on Operationalising Macroprudential Policy in the Banking Sector. Addendum: Macroprudential Leverage Ratios. European Systemic Risk Board, Frankfurt.
FSB. 2011. Policy Measures to Address Systemically Important Financial Institutions. Financial Stability Board, Basel.
Graham, J.R. 2006. A Review of Taxes and Corporate Finance. Foundations and Trends in Finance 1(7): 573-691.
Gropp, R. and F. Heider. 2009. The Determinants of Bank Capital Structure. Review of Finance 14(4): 587-622.
Heckemeyer, J. And R. de Mooij, R. 2013. Taxation and Corporate Debt: Are Banks any Different? IMF Working Paper 13/221.
Hemmelgarn, T. & Teichmann, D. 2014. Tax Reforms and the Capital Structure of Banks. International Tax and Public Finance 21(4): 645-693.
IMF. 2014. Global Financial Stability Report, April, Chapter 3: How big is the Implicit Subsidy for Banks Considered Too Important to fail? International Monetary Fund, Washington.
IMF. 2016. Tax Policy, Leverage and Macroeconomic Stability, IMF Staff Paper.
Kroszner, R. 2013. A Review of Bank Funding Cost Differentials. Published at Booth School of Business, University of Chicago.
Langedijk, S., G. Nicodeme, A. Pagano, and A. Rossi. 2014. Debt Bias in Corporate Taxation and the Costs of Banking Crises in the EU. European Commission Taxation Papers 50.
Lin and Flannery. 2013. Do personal taxes affect capital structure? Evidence from the 2003 tax cut. Journal of Financial Economics. 109(2): 549-565.
Luca and Tieman. 2016. Financial Sector Debt Bias. IMF Working paper 16/217.
Mooij, R. de and M. Keen, 2016. Debt, Taxes and Banks. Journal of Money, Credit and Banking 48(1), 5-33.
15
Mooij, R de and M.P. Devereux. 2011. An Applied Analysis of ACE and CBIT reforms in the EU. International Tax and Public Finance 18(1): 93-120.
Mooij, R. de, M. Keen and M. Orihare. 2014. Taxation, Bank Leverage and Financial Crises, Chapter 11 in Taxation and Regulation of the Financial Sector, de Mooij and Nicodeme (ed), MIT Press.)
Morrison, A. 2011. Systemic risks and the ‘Too-big-to-fail’ Problem. Oxford Review of Economic Policy 27(3): 498-516.
Myers, S.C. 1984. The Capital Structure Puzzle. Journal of Finance 39(3): 575-592.
Noss, J. and R. Sowerbutts. 2012. The Implicit Subsidy of Banks. Financial Stability Paper 15.
Schich, S. and S. Lindh, S. 2012. Implicit Guarantees for Bank Debt: Where do we Stand? OECD Journal: Financial Market Trends 1.
The Economist. 2015. Tax-free debt: The great distortion.
Ueda, K. and B. Weder di Mauro. 2013. Quantifying Structural Subsidy Values for Systemically Important Financial Institutions. Journal of Banking and Finance 37(10):3830-3842.
We thank, without implicating them, Jack Bekooij, Paul Hilbers, Mark Mink, Robert Mosch, Dirk Schoenmaker, two anonymous referees and participants at seminars at the Netherlands Economists Day, the Dutch Ministry of Finance and De Nederlandsche Bank. The usual disclaimer applies.
Jesse Groenewegen Rabobank; email: jesse.groenewegen@rabobank.nl
Peter Wierts De Nederlandsche Bank, VU Amsterdam, member of the ESRB Instruments Working Group and Assessment Team; email: peter.wierts@dnb.nl
Imprint and acknowlegements
© European Systemic Risk Board, 2017
Postal address 60640 Frankfurt am Main, Germany Telephone +49 69 1344 0 Website www.esrb.europa.eu
All rights reserved. Reproduction for educational and non-commercial purposes is permitted provided that the source is acknowledged.
Note: The views expressed in ESRB Working Papers are those of the authors and do not necessarily reflect the official stance of the ESRB, its member institutions, or the institutions to which the authors are affiliated.
ISSN 2467-0677 (online) ISBN 978-92-95210-40-0 (online) DOI 10.2849/49060 (online) EU catalogue No DT-AD-17-020-EN-N (online)